Problem
Research questions and friction points this paper is trying to address.
rainfall simulation
autonomous driving
physical controllability
multi-view consistency
3D Gaussian Splatting
GSRAIN: Physically Calibrated High-/Low-Frequency Rainfall Synthesis for 3D Gaussian Driving Scenes
🤖 AI Summary
Existing methods for simulating rain in autonomous driving scenarios suffer from limitations in physical controllability and multi-view consistency. This work proposes a rainfall synthesis framework based on 3D Gaussian splatting, which integrates a high-frequency raindrop model with a geometry-aware single-step diffusion model to generate low-frequency rain streaks and mist. By fusing these components, the method achieves physically calibrated, multi-view consistent, and intensity-controllable rain effects across a range of 0–13 mm/h—a first for 3D Gaussian-based scenes. Experiments demonstrate that the approach attains an FID of 149.09, outperforming CycleGAN-Turbo and WeatherEdit. Furthermore, object detection and closed-loop driving evaluations confirm its effectiveness in revealing performance degradation in perception and planning algorithms under adverse weather conditions.
📝 Abstract
Existing rainfall simulation methods for autonomous driving remain limited in physical controllability and multi-view consistency. This paper presents GSRAIN, a high-/low-frequency rainfall synthesis method for 3D Gaussian Splatting (3DGS) driving scenes. GSRAIN constructs a high-frequency raindrop model from measured rainfall data and generates low-frequency rainy appearance using a geometry-aware single-step diffusion model. The two effects are then fused in a unified 3DGS scene, enabling rainfall-intensity control over the range of 0--13~mm/h. The proposed method achieves a Fréchet Inception Distance (FID) of 149.09, outperforming CycleGAN-Turbo (155.71) and WeatherEdit (157.94). Object-detection and closed-loop driving experiments further show that the generated scenes expose scene-dependent performance changes of the evaluated algorithms under controllable rainfall. These results indicate that GSRAIN provides an effective approach for constructing physically controllable, repeatable, and closed-loop-compatible rainy-weather test scenes for autonomous driving.
Innovation
Methods, ideas, or system contributions that make the work stand out.
rainfall synthesis
3D Gaussian Splatting
physically calibrated
multi-view consistency
diffusion model
🔎 Similar Papers
No similar papers found.
💼 Related Jobs
